The objective of the proposed research is to assess the impact of intergenerational exchanges and filial expectations on caregiver and elder outcomes in daughter caregiver/ impaired elder dyads. The specific aims are to (1) refine multidimensional measures of intergenerational assistance and filial expectations, (2) fit a multiple-measure latent construct theoretical model that simultaneously estimates the direct and indirect effects of reciprocity on caregiver and elder well-being, (3) assess effects of changes in elder impairment, intergenerational exchanges and filial expectations on caregiver and elder well-being over an 18 month period, and (4) assess race and sex differences in the above processes. Data will be collected at T1 from 500 dyads of impaired elders and their primary caregiver daughters. The sample will be stratified by race and elder sex and screened from a sampling frame of 122,000 households in the Detroit MSA. Approximately 400 dyads will be reinterviewed 18 months later. This project is designed to extend previous research by (1) obtaining a sample of impaired elders and their daughter caregivers, (2) developing multidimensional measures of reciprocity and intergenerational expectations for both elders and caregivers, (3) assessing change over time in the effects of reciprocity on well-being and long-term care outcomes, and (4) exploring variability in the processes by race and sex.